Mapping-related service providers (e.g., map data providers, navigation service providers, etc.) face significant technical challenges when generating and updating geographical data for distribution to end users. These challenges are particularly acute in the context of real-time distribution and/or collection of geographical data (e.g., real-time traffic incident reports) to or from vehicles or other users of a transportation network. Because such geographical data or reports can be potentially numerous and their respective locations can be imprecise, service providers typically cluster the data to avoid potential duplication and/or ambiguity about a reported incident. However, performing global clustering of high volumes of dynamic geographical data can be resource constrained (e.g., with respect to computing resources, network bandwidth, etc.) to the point where real-time collection, processing, and/or distribution of the data may not feasible.